Bookmark and Share

Work in Progress

The Hewlett Foundation Blog




Working Toward a Data Revolution 

May 15, 2014 — By Rachel Quint

In her first Friday Note for the blog last fall, Ruth Levine wrote about the emerging data revolution. She described a “cacophony of voices, each with valid but distinct perspectives on the aims and methods of generating better data for development policy.” Since then, these voices have only grown louder, and a healthy (yet often disorganized) debate continues. Like the political revolutions of recent years, much of the conversation about the data revolution has taken place on social media. Across the blogosphere, brave development specialists have provided suggestions, ideas, and warnings about the direction this revolution is headed.

The enthusiasm surrounding the data revolution is a testament to the need for a new approach to data. A successful revolution would mean more and better data about what is happening in developing countries and greater availability of that data. This could allow citizens to hold governments accountable for delivering on development promises.

What are these revolutionaries trying to achieve? The online conversations point to three main data challenges:

 (1) Addressing data gaps—collecting more data (increasing the quantity of data) targeted for specific purposes, including the use of non-traditional data collection methods

 Some, who view the data revolution as an element of the post-2015 development agenda, are recommending a large-scale survey to monitor countries’ progress against the to-be-identified post-2015 framework goals, targets, and indicators. A large-scale data collection process would create comparable, nationally representative data for the indicators with enough detail in data collection to permit disaggregation by gender, and even geography, income, or disability. There are also calls to utilize technology to collect more data at a lower cost, including collection of big data (using methods that utilize secondary data, such as cell phone records or internet use, or crowdsourcing methods, such as SMS surveys).

(2) Improving data quality—recognizing chronic shortcomings in developing countries’ data systems; improving data quality and the processes for collecting it

Improving data quality would come mainly through capacity support to national statistics offices, which are the institutions that collect the largest quantity of human development indicators. This could take the form of increased funding for national statistic offices and domestic civil societies who are already collecting data, but require capacity support, or greater support for the collection of civil registration and vital statistics (such as birth and death records).

(3) Optimizing and expanding use of data—ensuring that data is effectively used by those within and outside the official statistics infrastructure

Building on the momentum created by the Open Data Charter, which declares that government data is a public good, many are focused on access to data. They are also highlighting the role that data should play in government accountability, including providing financial and revenue data to citizens to curb corruption and increase the responsiveness to citizens’ needs.

Who will lead the Revolution? The final piece of the puzzle is the design of a Global Partnership across institutions, countries, and civil society to support a successful Data Revolution. Molly Elgin-Cossart, a Senior Fellow at the Center for American Progress and one of the HLP report authors, has provided some excellent commentary on the potential structure and purpose of the Partnership. Elgin-Cossart suggests that the Partnership could take one of several structures, including: a single agency with a broad representative board; a network of partners supported by a small but nimble support team; a hub of thematic partnerships that focus on generating data for their specific subject-matter (e.g., education, food security); or a loose network united by common principles and objectives.

Adopting any of these formats for building a data revolution will be a considerable challenge. Clarity around the need for change, and even agreement about ultimate aims are no guarantee of success. But no one ever said that revolutionary change was easy.